In the first five years of work, we showed that the study of convenience samples of treated alcoholics embody a bias, called `Berkson's Fallacy', in that treated individuals with AD differ from treatment naive individuals with AD in the general population, who comprise the majority of alcoholics. We studied 21-50 year old actively drinking treatment naive AD (TNAD) and age and gender comparable controls, and compared them to our other ongoing study of long-term abstinent treated AD (ATAD). We showed that with regard to alcohol intake in the initial period after starting abusive drinking, TNAD were not simply treated AD observed earlier in the progression of the disease, but rather comprise a subpopulation with lower initial heavy alcohol consumption than ATAD. Regarding psychiatric comorbidity, we showed that TNAD had more psychiatric disorder than non-alcoholic controls (NAC), but less psychiatric disorder that ATAD. We also showed that TNAD have more normal brain structure on MRI than ATAD, and that TNAD are cognitively normal despite active heavy drinking. In the current proposal, we will study TNAD 35--60 years of age, most of whom would have drank abusively for longer periods (and have a much higher total lifetime alcohol intake) than our former TNAD. The major goals of the proposed research are: 1) to determine whether 35-60 year old TNAD differ in brain structure and function from age-comparable NAC, and from both short- and long-term ATAD (to be available from our pending project);2) to determine whether 35-60 year old TNAD differ from NAC and from both short- and long-term ATAD in the presence and severity of comorbid mood, anxiety, and externalizing disorder diagnoses, symptoms and traits;3) to determine the factors associated with impairments in brain structure and function in TNAD (including presence and severity of comorbid mood, anxiety, and externalizing disorder diagnoses, symptoms and traits, premorbid brain size as a measure of functional reserve (as indexed by intracranial vault volume on MRI), the pattern and severity of drinking, the frequency of withdrawal symptoms (including severity), and predisposing factors such as the family history of alcohol problems or electrophysiological and behavioral end phenotypic alcoholism markers));4) to determine whether men and women differ in these comparisons;and 5) to determine the effect of concomitant drug abuse/dependence on these comparisons. PUBLIC HEALTH RELEVANCE We have previously shown that 30-year old treatment-naive alcohol dependent individuals are intermediate between non-alcoholic controls and treated alcoholics. In the current proposal, we plan to continue this work, studying 35 - 60 year old treatment naive alcohol dependent individuals. Such individuals should have a much greater lifetime alcohol burden than the 30 year old samples studied previously. Additionally, we are including a sample of treatment-naive drug and alcohol dependent individuals, to provide greater generalizabilty to the U.S. population, and test the hypothesis that concomitant drug dependence is associated with more severe AD, greater psychiatric comorbidity, and greater cognitive and brain structural morbidity in treatment naive AD. These middle-aged TNAD comprise about 40% of all alcoholics, and the great majority never seek treatment. Since most studies of alcoholism in the U.S. use convenience samples of alcoholics in treatment (a minority of all alcoholics), the proposed studies are essential to understand the impact of AD on the US population.